Lines Matching defs:start
989 * close off a transaction and start a new one between the ext4_get_block()
990 * and the commit_write(). So doing the jbd2_journal_start at the start of
1168 * is being written back. So grab it first before we start
1318 * ordering of page lock and transaction start for journaling
1367 unsigned start, size;
1369 start = max(from, block_start);
1370 size = min(to, block_end) - start;
1372 zero_user(page, start, size);
1564 ext4_lblk_t start, last;
1565 start = index << (PAGE_SHIFT - inode->i_blkbits);
1573 ext4_es_remove_extent(inode, start, last - start + 1);
1899 * We need to release the page lock before we start the
1969 * Note that we don't need to start a transaction unless we're journaling data
1975 * cannot start transaction directly because transaction start ranks above page
2207 * @bh - buffer we should start processing from
2351 pgoff_t start, end;
2357 start = mpd->map.m_lblk >> bpp_bits;
2359 lblk = start << bpp_bits;
2363 while (start <= end) {
2365 &start, end);
2576 * IO immediately. When we find a page which isn't mapped we start accumulating
2837 /* start a new transaction */
3082 unsigned long start, end;
3096 start = pos & (PAGE_SIZE - 1);
3097 end = start + copied - 1;
3134 * also start triggering a write of the data blocks, which is
3160 * which will map the blocks, and start the I/O, but not
3472 * don't need to start a journal txn and we can directly return
3884 ext4_fsblk_t start, end;
3891 start = lstart >> sb->s_blocksize_bits;
3895 if (start == end &&
3901 /* Handle partial zero out on the start of the range */
4515 int i, start;
4517 start = inode_offset & ~(inodes_per_block - 1);
4533 for (i = start; i < start + inodes_per_block; i++) {
4540 if (i == start + inodes_per_block) {
5389 * we start writeout of all the dirty pages which are being truncated.
5987 * we start and wait on commits.